Transportation vehicle control with phoneme generation
AssigneeVolkswagen Aktiengesellschaft
InventorsOkko Buss, Mark Pleschka
Life of this patent
Estimating the term…
Abstract
A transportation vehicle having a navigation system and an operating system connected to the navigation system for data transmission via a bus system. The transportation vehicle has a microphone and includes a phoneme generation module for generating phonemes from an acoustic voice signal or the output signal of the microphone; the phonemes are part of a predefined selection of exclusively monosyllabic phonemes; and a phoneme-to-grapheme module for generating inputs to operate the transportation vehicle based on monosyllabic phonemes generated by the phoneme generation module.
Claims8
A transportation vehicle comprising:
- a navigation system;
- a user input control connected to the navigation system via a bus system for data interchange purposes,
- a microphone;
- a phoneme generation module that includes a statistical language model for generating phonemes from a voice signal and/or the output signal of the microphone, wherein the phonemes are part of a prescribed selection of exclusively monosyllabic phonemes;
- and
- a phoneme-to-grapheme module that includes a recursive neural network, the recursive neural network comprising between 2 and 4 intermediate layers and a recall layer having a dimension of between 10 and 30, the recursive neural network generating inputs for controlling the transportation vehicle based on a succession of monosyllabic phonemes generated by the phoneme generation module, wherein the phoneme-to-grapheme module is trained substantially exclusively using entries associated in pairs, which entries comprise a monosyllabic phoneme that has a respective associated monosyllabic term.
A method for manufacturing a transportation vehicle that includes a navigation system, a user input control connected to the navigation system via a bus system for data interchange purposes, a microphone, a phoneme generation module that includes a statistical language model for generating phonemes from a voice signal and/or the output signal of the microphone, wherein the phonemes are part of a prescribed selection of exclusively monosyllabic phonemes, and a phoneme-to-grapheme module that includes a recursive neural network, the recursive neural network comprising between 2 and 4 intermediate layers and a recall layer having a dimension of between 10 and 30, the recursive neural network generating inputs for controlling the transportation vehicle based on a succession of monosyllabic phonemes generated by the phoneme generation module, wherein the method comprises:
- providing a first database of inputs or commands for controlling functions of the transportation vehicle;
- generating a second database that comprises exclusively monosyllabic phonemes;
- training the phoneme generation module by using the first database;
- training the phoneme-to-grapheme module by using the second database, wherein the phoneme-to-grapheme module is trained substantially exclusively using entries associated in pairs, which entries comprise a monosyllabic phoneme that has a respective associated monosyllabic term;
- connecting the output of the phoneme generation module to the input of the phoneme-to-grapheme module for data interchange purposes;
- and
- implementing the phoneme generation module and the phoneme-to-grapheme module in a transportation vehicle.
Dependencies are inferred from the claim text.